PulseAugur
中
实时 23:24:12
English(EN) How many tokens does a web page cost your LLM? We measured 20 real pages

LLM网页token成本:原始HTML效率低下,Markdown节省98%

一项研究测量了检索增强生成(RAG)管道和AI代理处理网页的token成本。研究结果表明,原始HTML网页的token效率非常低下,中位数为约128,000个token。将这些网页转换为面向LLM的Markdown格式,在仅提取主要内容并进行进一步清理后,与原始HTML相比,token数量中位数减少了98%,与整页文本相比减少了63%。研究强调,主要内容提取是减少token使用的最有效步骤,节省量因网页结构而异。 AI

影响 为LLM优化网页处理可以显著降低RAG系统的成本并提高检索准确性。

排序理由 该条目展示了一项关于LLM处理网页的token效率测量研究的发现。[lever_c_demoted from research: ic=1 ai=1.0]

在 dev.to — LLM tag 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

LLM网页token成本:原始HTML效率低下,Markdown节省98%

本文如何被排名

Signal score
26 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该条目展示了一项关于LLM处理网页的token效率测量研究的发现。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
infra, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Leif Albrand ·

    一个网页会消耗你的LLM多少token?我们测量了20个真实网页

    <p><em>Raw HTML vs. full-page text vs. main-content Markdown — median, range, and the cases where the saving is small.</em></p> <p>If you build RAG pipelines or agents that read the web, every page you fetch ends up as input tokens: in the embedding step, in the context window, o…